Bosch Factory Lambda Tester Tool - PeachParts Mercedes-Benz Forum

tybee380sl · #1 08-07-2009, 08:43 PM

Can anyone comment on the circuitry of the Bosch Lambda tester or any similar tools that are still available today? How many pins are on the x11 mating connector and does anybody know which pins are utilized for the different settings? (100, 100IR, etc)

It almost seems to me like it is only an analog dwell meter with a switch on it to change the internal circuitry to measure the dwell/duty cycle across different pins of the x11 connector depending on model year of the vehicle. The reason I question this is because I have read conflicting information regarding duty cycle measurement (air/fuel mixture settings) on K Jet w/Lambda SLs (1981-1985) . Some people measure voltage, duty cycle %, or dwell across pins #3 & 2 and some measure across #6 & 3. All with varying degrees of success. It seems like some people get lucky and have a friend or go to a shop that has one of these older tools and they can get their air/fuel mix right on.

Strife · #2 08-08-2009, 08:18 PM

The signal (can't remember which pin) is a ~12 volt, ~100hz square wave with (nominally) a 50% duty cycle (meaning, it's "high" 50% and "low" 50%.If it were high 90% of the time, that would be 90% (not 10%). The dwell meter would comparitively show you something, but it's hard to say where 50% is. If you operate a dwell meter at a 4 cyl position, when you run 0-90 on the scale, that would be close to 0-100% duty cycle.

I occasionally get the urge to make one with a little microcontroller and a few LED displays. You can buy one (they are sold for fuel injectors, kids put them on their dashboards). The firmware might be a little tricky because you wouldn't want the display to update too quickly (bouncing around all over the place), maybe some kind of average per 1/3 second would be good.

ctaylor738 · #3 08-09-2009, 09:20 AM

I've had success with my Sears analog dwell meter/volt meter on Pin 3. For an 8 cylinder with a 45 degree scale, I set it around 22 degrees. But you could just as well use the 4 or 6 cylinder scale and set to 45 or 30 degrees, respectively.

tybee380sl · #4 08-09-2009, 07:34 PM

On my 1984 380sl, I've hooked up my dwell meter to the x11 (red to #3, black to #2), disconnected the O2 sensor, grounded the oil temp sensor, and I measure 30 degrees on a 4cyl scale. My understanding is that I should be measuring around 45 degrees indicating about a 50% duty cycle. This leads me to believe I am not measuring the dwell/duty cycle correctly or the Lambda controller is shot. I have a spare controller and I get the same readings with that. I'm thinking I've missed something fundamental here...

I'm wondering what is so unique about the factory tool that requires different settings for K Jet Lambda (100), KE Jet and up (100IR) and helps some people get their air/fuel mix set correctly that they were unable to accomplish with a DMM duty cycle meter or dwell meter.

Example: Post #10
Can't get Lambda leaner, 1983 380SL

Strife · #5 08-09-2009, 11:24 PM

In my post, I stated that "ON" would be 100% (which would be the classic definition of a "duty cycle"), actually, on this system, it works the opposite way; 30% HIGH, and 70% LOW, would be (in this system) a 70% duty cycle (it just depends on whether you consider "high" or "low" true).

ctaylor738 · #6 08-10-2009, 07:51 AM

But if you ground the wire to the oil temp sensor, then you throw the system into warm-up/acceleration mode, which is a fixed 70% duty cycle. If you unground the wire you should see it cycling around 50% of whatever scale you are reading.

tybee380sl · #7 08-10-2009, 08:49 AM

I was grounding the oil temp sensor to keep the car in warm-up and I was looking for a fixed 60/40 duty cycle. This is the first time I have really seen mention of a 70/30 warm-up for these cars. Just to confirm, when I unground the oil temp sensor, leaving the O2 sensor disconnected, should I see about a 50% fixed duty cycle? Or would it be different for K Jet w/ Lambda systems?

This is what I measure:
Oil temp sensor grounded, O2 disconnected: 70/30 duty cycle (27 degrees of dwell on a 4cyl scale)

Oil temp sensor not grounded, O2 disconnected: 75/25 duty cycle (22.5 degrees of dwell on a 4cyl scale)

ctaylor738 · #8 08-10-2009, 09:56 AM

Sorry, I glazed over the oxygen sensor disconnect. I would think that the brain would see that a fault, and go to some sort of fixed ratio. Maybe that's what the 70/30 is. But why it would go to 75/25 with the temp wire ungrounded escapes me.

With a warm engine, sensor connected, temp wire ungrounded, mixture set properly, no vacuum leaks or false air, and the planets aligned, you should see a fluctuation around 50%.

tybee380sl · #9 08-10-2009, 12:59 PM

On K Jet with Lambda systems, what controls the fixed fuel mixture post warmup and prior to closed loop operation? The oil temp sensor opens at about 60F and the O2 sensor doesn't send reliable information back to the controller unless its somewhere above 400F.

I know the throttle position sensor at WOT kicks the K Jet back to the same fixed duty cycle (70/30 I guess) but I can't tell from the wiring diagrams what other input could be influencing the fixed fuel mixture while it is still open loop. I've checked the throttle sensor according to the Mercedes service procedure (resistance checks at different positions) and it looks to be ok.

ctaylor738 · #10 08-10-2009, 01:48 PM

My understanding is that it goes into closed loop when the oil temp switch opens. If the sensor was not hot enough, it would send a lower voltage signal which would cause the brain to richen the mix, which would be OK since the engine would still be warming up.

I dimly recall that when my various cars went into closed loop, the duty cycle went rich briefly, and then stabilized.

From a conversation I had with a K-Jet expert a while ago, there are four states for the system:

- fixed rich (warm-up, acceleration)
- fixed lean (high vacuum, closed throttle as in coasting)
- variable (closed loop)
- fixed ? (no sensor input)

tybee380sl · #11 08-10-2009, 09:21 PM

So maybe I need to give it a chance to stabilize and then see what happens once warm and in closed loop. Thanks.

tiehacker · #12 08-25-2009, 06:15 PM

I use a common analog voltmeter. Connect the negative lead to #3 pin of the diagnostic plug and the positive lead to the positive side of the battery. Disconnect O2 sensor and note the voltage reading. The needle should be steady. Reconnect the oxy sensor. The needle should vibrate. If the vibrations are centered on the noted voltage all is well. If not then adjustment is neccessary. Good luck Tiehacker

tybee380sl · #13 08-25-2009, 08:34 PM

Tiehacker, thanks for the reply. I've tried using an analog multimeter, a digital multimeter, and an analog dwell meter to measure the duty cycle. The best I can come up with is about 70/30% (depending on perspective) with the O2 sensor disconnected and all the aux. sensors (oil temp, throttle position sensor, idle control, etc., basically everything connected to pins 6 & 7 of the lambda controller) bypassed to make the car run open loop. Even closed loop (warm and O2 connected), duty cyle fluctuates but hovers around 75/25%. I can adjust the mixture to about 58/42% at which time it runs but if I shut down the engine, it will not restart until I richen the mixture back up. There must be something unique about the 83-85 K Jet w/lambda cars that the convention duty cycle measurements at pins #3 & 2 of the X11 connector are not the same as other years. The factory Mercedes service manual indicates a different technique for measuring the duty cycle on the 83-85 cars (match idle and 2500rpm measurements within 10% or so) which i don't see being as precise. I was just wondering what is so unique about the factory lambda tool that makes it functional on all model year sl's.

ctaylor738 · #14 08-26-2009, 10:41 AM

This is at least the third 380 with this problem. One was my '85.

Same exact symptoms. As you leaned the mixture toward 50/50, the engine would go rough and not want to start. Leave it at 70/30 or so and it was fine. This one ended up a tie. It had perfect system and control pressure. New injectors seals and tested injectors, I replaced the brain, the oxygen sensor, the throttle switch, the vacuum strainer, and finally the fuel distributor. Everything made it a little better but did not solve the problem.

I ended up connecting my meter between the oxygen sensor and ground with a jumper so that I could measure the voltage. I set the mixture so that it was between .45 and .5 volts and hooked it back up. At this point, I would get a 70/30-ish reading.

Starting means very little airflow over the plate and hence very little deflection hence very little fuel to the injectors. With a warm engine and no fuel from the cold start valve, the mixture setting is critical to getting enough fuel to start. So the the basic mixture needs to be set rich enough to allow the car to start, and it seems to be at a point where, when the car is warmed up, the brain will try to lean it.

One thing that helped was to adjust the warm-up regulator to the minimum warm control pressure of 3.4 bars. This increases the amount of fuel at a low deflection situation.

I personally think that the MB tool is really an analog duty cycle meter. But I have never been able to completely explain what happens in this situation, so maybe you are on to something. It would be interesting to see what you get from the MB tool.